Friday, 1 September 2017

Better Drugs, Faster: The potential of AI-powered Humans

Artificial Intelligence

AI – Reduce Time in Development of New Drugs & Cost

Tech companies have stated that scientists working on with artificial intelligence could reduce the time it tends to take in the development of new drugs and significantly also the cost. Increasing pharmaceutical drugs seems to be a very expensive and time consuming affair.

As discovered by AstraZeneca recently, disappointing drug trials could knock millions off your stock market value in a blaze. Hence the sooner we tend to identify promising molecules before it can be turned into viable drugs, the better.

 This is the reason why pharmaceutical companies like GlaxoSmithKline – GSK, Merck, Sanofi and Johnson & Johnson are now turning to artificial intelligence – AI to help them. Prof Andrea Hopkins is the chief executive of Exscientia which is an AI-based drug discovery firm that had recently signed a £33m deal with GSK.

As per Prof Hopkins, AI as well as human beings functioning collectively in a centaur team could be supportive in identifying candidate molecules in a quarter of the usual time and at a quarter of the cost. As per Prof Hopkins’s belief the centaur in Greek mythology is said to be half humans, half horse and seems to be very powerful as well as fast and AI has been providing scientists with such extra power.
 

`In Silico” – Medical Term for Research

 
Successful drug discovery is dependent on actual comprehension on how an ailment tends to affect our biological systems according to global life sciences industry leader at consultancy firm EY, Pamela Spence.

She clarifies that once it is identified scientists then search for molecule which can selectively interact with this target and reverse that disruption or slow its impact, a `hit molecule. Scientists often tend to discuss about a disease as the target and the molecule as a weapon being fired at it.

 However, the procedure of drug discovery, which is usually carried out by small groups of scientists carefully investing each potential target and hit molecule in the hope of discovering a winner, seems to be an enormously time-consuming method which also seems to have a very high failure rate.

She states that, considering artificial intelligence is like having a research assistant who has the potential of solving problems through systematic and relentless search at incredible speeds. She further added that what could work and equally essentially what could not work can be identified initially by the AI supercomputer `in silico’. This is said to be a medical term for research that has been done by computer as against `in vitro’ – think test tubes and `in vivo’ – testing on animals and humans.
 

Human Clinical Trials – Massive Bulks of Drug Discovery Cost

 
Conducting human clinical trials accounts for the massive bulks of drug discovery cost, and the quicker we tend to identify when something is not moving in the right direction the less money could be saved. Ms Spence had stated that then the physical testing can be conducted on a smaller number of possible new medicines and a much higher success rate can be achieved.

The artificial intelligence process of Exscientia heads masses of data from the structure of diseases to the ability of prevailing drugs, from peer-reviewed studies to observations of slides under a microscope. All these possibilities have been tapered down in a process where Prof Hopkins relates to natural selection.

He commented that they are not attempting to rule out the uncertainty, this is messy, dirty data and these are very interesting analogies between how human creativity tends to work and growth. The purpose is to come up with small molecules as candidates for around 10 disease-connected targets which could then be put through medical examinations.

Prof Hopkins, who is also chair of medicinal informatics at the UK’s Dundee University, had informed that every pill one may make could cost pence to manufacture though it is actually a precision-engineered product. He further added that there is almost infinite number of other molecules which could have been. One needs to make decisions as to which one could be safe and effective.

 

Cost – Target To Clinic

 
Most do not lead to anything. The artificial intelligence driven approach also tends to make it easy to come up with molecules which tend to have two diverse goals. For instance, a cancer drug tends to also improve the immune system and is also said to confront the disease.

GSK intends getting behind the idea and recently has set up a discovery performance unit directed on enhancing drug discovery via the use of `in silico’ technology comprising of artificial intelligence, machine learning together with deep learning.

The initiative is being headed by John Baldoni, head of R&D of GSK. He has commented that they have various amounts of these deals which they intend putting in place and the one with Exscientia is possibly the one which is furthest along.

 However they have a few others in flow as well a few internal projects themselves. He has also informed that the cost of discovery from target to launch is around $1.7bn (£1.3bn). The cost related here is from target to clinic which is around 33% of that and it would take about five and a half years. He added that their goal was to reduce that to one year and reduce the cost proportionate along with that.
 

AI-Based Drug Discovery – Effective/Cheaper Drugs – Quicker

 
Artificial intelligence has also been making its way into other characteristics of the drug discovery process. For instance, Benevolent AI tends to utilise natural language process in order to select through published literature like chemical libraries, medical databases together with scientific papers in order to come to some decisions regarding the possible new drug candidates.

One of the its candidates for a drug to treat motor neurone disease which is also known as Amyotrophic Lateral Sclerosis – ALS, was earlier this year, found to prevent the death of motor neurones in cells taken from real patients and deferred the onset of the disease in animals.

The founder and chairman of Benevolent AI, Ken Mulvaney had commented that they were extremely encouraged by these discoveries. Patients need to be encouraged also. The artificial intelligence-based drug discovery tends to be promising in bringing about more effective, together with cheaper drugs on to the market in a much quicker manner.

No comments:

Post a Comment

Note: only a member of this blog may post a comment.